In Research Component 4, we will develop and clinically test endomicroscopic Collagen I (Coll) fiber imaging with an innovative ultra-compact fiber-optic biopsy needle-compatible endomicroscope that is optimized for second harmonic generation (SHG) microscopic detection of Coll fibers. We will integrate this endomicroscope with a routinely used clinical 14-gauge biopsy needle for SHG Coll imaging in preclinical and clinical studies. Coll fibers will be detected in primary breast tumors at the time of biopsy to predict lymph node involvement in breast cancer patients. The assessment of lymph node involvement, which is the most important predictor of prognosis, early on in the clinical management of breast cancer patients, would be of significant advantage in selecting treatment options at the early time of the initial breast biopsy. Coll fibers are an important component of the extracellular matrix (ECM) in breast tumors, and increased stromal Coll fibers facilitate breast tumor formation, invasion, and metastasis. High Coll fiber density in primary tumors promotes metastasis as cancer cells migrate along dense CoH fiber avenues within metastatic primary tumors to form metastatic nodules. We will therefore evaluate Coll fiber density as an imaging biomarker to predict tumor aggressiveness and lymph node metastasis. Coll fibers will be detected by optical SHG microscopy, a highly sensitive and noninvasive technique for imaging intrinsic signal from Coll fibers. We observed an increased Coll fiber density in lymph node positive (LN+) compared to lymph node negative (LN-) patients using ex vivo primary tumor specimens from breast cancer patients. Molecular optical endomicroscopic SHG imaging of Coll fibers, which directly reveals ECM integrity and structure, may serve as a surrogate marker to predict metastasis. We will test this hypothesis in three Aims. In Aim 1, we will quantitatively characterize SHG-detected Coll fiber signatures in primary tumor specimens from breast cancer patients and evaluate if the quantitative Coll fiber parameters 'fiber distance' and 'fiber volume' predict lymph node status. In Aim 2, we will develop, test, and optimize an ultra-compact fiber-optic SHG endomicroscope to detect Coll fibers. In Aim 3, we will perform the first clinical feasibility study with this fiber-optic SHG endomicroscope during standard ultrasound-guided breast biopsy procedures and thereby obtain SHG Coll images that can help predict lymph node metastasis. In Aims 1 and 3, we will also correlate Coll fiber density with standard breast cancer prognostic markers to determine the ability of CoH fiber density to uniquely select a population of patients that are at risk for lymph node involvement and potentially systemic disease. On a wider perspective, the device being built will also be useful for future applications such as detecting response to stromal depletion therapies that improve drug delivery in desmoplasmic tumors, and predicting drug delivery.